Role of corpus callosum in interhemispheric coherent activity during sleep

Role of corpus callosum in interhemispheric coherent activity during sleep

M. Corsi-Cabrera a,*, R. Ondarza b, V. Martınez-Gutierrez a, Y. del Rıo-Portilla a,

M.A. Guevara c, J. Ramos-Loyo c

a Facultad de Psicologıa, Posgrado, Universidad Nacional Autonoma de Mexico, Mexico City, Mexicob Instituto Nacional de Neurologıa y Neurocirugıa MVS, Mexico, DF, Mexicoc Instituto de Neurociencias, Universidad de Guadalajara, Jalisco, Mexico

Accepted 21 May 2006

Abstract

Objective: To investigate to what extent the increase in interhemispheric coherent activity observed from wakefulness to sleep depends on the

integrity of the corpus callosum (CC).

Methods: Interhemispheric coherent activity was analyzed in two epileptic patients selected for callosotomy because of multifocal refractory

epilepsy, before and 4 months after callosotomy. One patient underwent complete callosotomy and another was subjected to callosotomy of

the anterior 2/3, which offered the possibility of comparing the role of the CC in the coherent activity increase from wakefulness to sleep,

between anterior regions with interrupted CC communication (in the two patients) and posterior regions with intact communication (in one of

them). Results were compared with a group of normal subjects.

Results: Both patients showed increased coherent activity from wakefulness to sleep after surgery.

Conclusions: Results demonstrate that interhemispheric coherent activity, despite an attenuation after surgery, is higher during SWS than

during wakefulness after sectioning the CC; however, they have to be taken with caution because they come from two patients only.

Significance: Present results show that the increase in coherent activity during sleep does not depend exclusively on callosal integrity but also

on state-dependent influences from sleep-promoting mechanisms, probably spread throughout the thalamo–cortical network.

q 2006 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

Keywords: Epilepsy; Callosotomy; Corpus Callosum; EEG coherence; EEG correlation; Sleep

1. Introduction

In the last decades, the importance of coherent activity

between neuronal populations of different brain regions for

cognitive processes has been stressed, and this cooperative

activity has been proposed as a binding mechanisms serving

to integrate disperse information in cortical regions into a

unified experience (Crick, 1994; Singer, 1999; Gray, 1999;

Edelman and Tononi, 2000). In addition to the stimulus

time-locked increase of coherent activity between the

cortical regions directly involved in processing a specific

1388-2457/$30.00 q 2006 International Federation of Clinical Neurophysiology.

doi:10.1016/j.clinph.2006.05.008

* Corresponding author. Address: Laboratorio de Sueno, Facultad de

Psicologıa, Universidad Nacional Autonoma de Mexico, Av. Universidad

3004, Copilco-Universidad, Mexico City DF 04510, Mexico. Tel.:C52 55

56 22 22 51/55 68 95 14.

E-mail address: [email protected] (M. Corsi-Cabrera).

task, the basal level of coherent activity between homo-

logous regions of the left and right hemisphere undergoes

dramatic changes with sleep. In independent groups of

subjects, we found a significant increase in interhemispheric

coherent activity of slow Hertz and spindle frequencies from

wakefulness (W) to stage 2, and stage 4 of slow wave sleep

(Corsi-Cabrera et al., 1987, 1996; Guevara et al., 1995;

Perez-Garci et al., 2001). The increase of coherent activity

during sleep has been confirmed by other studies (Nielsen et

al., 1990; Achermann and Borbely, 1998). The meaning of

the increase in interhemispheric coherent activity during

sleep is not fully understood yet but it suggests a

reorganization of functional relationships among cortical

and thalamic areas that may be relevant for the reprocessing

of information; the changes in intrinsic excitability of

cortical neurons and coherent oscillations in the cortex and

thalamus that take place during sleep may provide an

Clinical Neurophysiology 117 (2006) 1826–1835

www.elsevier.com/locate/clinph

Published by Elsevier Ireland Ltd. All rights reserved.

M. Corsi-Cabrera et al. / Clinical Neurophysiology 117 (2006) 1826–1835 1827

optimal milieu for reinforcing plastic connections (Steriade,

1998). In fact, the pattern of correlated unitary activity

observed during acquisition of a spatial task is replicated

during subsequent sleep in rat hippocampus (Wilson and

McNaughton, 1994), and the same cortical areas involved in

the training of memory tasks during wakefulness increase

their metabolic activity and (Maquet et al., 2000; Laureys

et al., 2001) become coherent (Gais and Born, 2004; Molle

et al., 2004) during sleep in man. Furthermore, the increase

during sleep in coherent activity has an impact on

subsequent wakefulness (Corsi-Cabrera et al., 1989),

which indicates the importance of investigating the

underlying mechanisms.

Reciprocal connections between two neuronal

assemblies have been proposed to underlie coherent activity

between them (Tononi et al., 1992). Among the cerebral

commissures, the corpus callosum (CC) is the major

pathway for the transfer of electrical activity from one

hemisphere to the other (McIntyre, 1995) and thus probably

also for the communication and integration of sensory and

cognitive information between cortical areas of the two

hemispheres (Schwartz, 1995); however, the extent to which

the CC connections are the sole mechanism responsible for

the increase in coherent activity between cortical regions

during sleep is still a matter of debate. Case reports of

bilateral asynchronous sleep spindles (Kuks et al., 1987;

Fuchigami et al., 1996), lower values of interhemispheric

coherence during REM and NREM sleep in congenital

agenesis of the corpus callosum (Nielsen et al., 1993) and

after partial callosotomy in two patients (Montplaisir et al.,

1990) are reported in the literature. Lower interhemispheric

coherence in all vigilance states over the entire 0.5–25 Hz in

NREM sleep and over all frequencies except 3–7 Hz in

REM sleep has also been recently reported in mice with

congenital callosal dysgenesis, (Vyazovskiy et al., 2004).

However, other case studies have not found differences in

interhemispheric coherent activity during wakefulness after

callosotomy (Corsi-Cabrera et al., 1995) or after interrup-

tion of cortico-cortical connections during sleep in the cat

(Contreras et al., 1996). This evidence together with the fact

that interhemispheric coherent activity is increased during

sleep between most cortical areas in specific frequencies of

the spectrum, suggests that not only interhemispheric

connectivity, but also state-dependent influences, such as

sleep promoting mechanisms, may be involved in the

general level of coherent activity during sleep.

The aim of the present study is to investigate to what

extent the increase in coherent activity during sleep

compared to wakefulness depends on corpus callosum

integrity. With this purpose, interhemispheric coherent

activity was analyzed in two epileptic patients selected as

candidates for CC surgery because of multifocal refractory

epilepsy. The CC has been considered the major pathway

involved in the bilateral spread of epileptic discharges

(Schwartz, 1995; Wada, 1995). Its role in the bilateral

spread and synchronization of epileptic seizures has been

extensively analyzed in animals (McIntyre, 1995), and in

clinical studies (Marino et al., 1995). For this reason, CC

section has become a surgical target to arrest the interhemi-

spheric spread of propagating seizures (Williamson, 1995)

providing the opportunity to study the role of the CC in the

sleep-related increase in coherent activity in the human

brain.

2. Method

We studied two right-handed male patients from the

Instituto Nacional de Neurologıa y Neurocirugıa MVS

(National Neurology and Neurosurgery Institute, MVS)

who were considered suitable for callosotomy surgery due

to medically intractable epilepsy, and because clinical

results, EEG recordings, and neuroimaging evaluations

failed to provide localizing information needed to perform

resective surgery. They gave their written informed consent

to participate in the EEG study. The EEG recordings were

conducted in accordance with the Declaration of Helsinki.

The patients spent two consecutive nights at the sleep

laboratory before and 4 months after CC-surgery. The

clinical, pharmacological and sleep profile of the patients

are shown in Table 1. Trans-operative ECoG in the patient

RC allowed to make the decision to continue with a

complete callosotomy in the same surgery. They were

selected because one of them (RC) underwent complete

callosotomy (right fronto-parietal craniotomy) and the other

(AS) anterior 2/3 callosal section only (right anterior

craniotomy), providing the possibility to compare the role

of the CC in the increase in coherent activity from W to

SWS, between cortical regions with interrupted (anterior

regions in the two patients) and with intact CC communi-

cation (posterior regions in one of them). Results were

compared with a group of normal subjects of the same sex

and similar age range consisting of eight adult right-handed

male volunteers, aged 23–30 years. Significant differences

in interhemispheric coherent activity between stages 2, 4,

REM sleep and wakefulness of these subjects have been

reported elsewhere (Perez-Garci et al., 2001; Corsi-Cabrera

et al., 2006) and served as parameters to compare sleep and

wakefulness in the patients.

EEG activity from the 19 electrodes of the 10–20

International System referred to ipsilateral earlobes was

recorded using a Grass model 8-20E polygraph with EEG

filters set at 1 and 70 Hz. Ipsilateral ear lobes were chosen as

reference to avoid as far as possible the contamination of

coherent activity between the two hemispheres by a

common signal. Our results from ipsilateral recordings are

congruent with findings from studies with bipolar deri-

vations (Achermann and Borbely, 1998; Duckrow and

Zaveri, 2005). Standard electrooculogram and electromyo-

gram activity was also recorded to identify sleep stages.

Impedance for the EEG electrodes was kept below 10 kU.Polysomnographic activity during spontaneous sleep on the

Table 1

Pre- and post-surgical clinical pharmacological and sleep profiles of the two patients

Patient Pre-surgical condition Post-surgical condition

RC AS RC AS

Age 38 26

Type of crisis AC AC CPC AC

CPCCSGCPC CPC SPC

SPC

TC

Frequency of crisis 20–28/month 32/month 2/month 18/month

Antiepileptic drugs PRM 875 mg/day PRM 750 mg/day PRM 750 mg/day PRM 750 mg/day

CBZ 1200 mg/day CBZ 600 mg/day CBZ 600 mg/day

PHT 350 mg/day PHT 300 mg/day PHT 300 mg/day PHT 300 mg/day

TST (min) 448.5 488.5 453.5 461

S4/TST (%) 0.5 6.7 2.9 11.7

S2/TST (%) 53.1 55.15 56.0 46.5

AC, atonic crisis; TC, tonic crisis; CPC, complex partial crisis; SPC, simples partial crisis; SGCPC, secondary generalized complex partial crisis; PRM,

primidone; CBZ, Carbamazepine; PHT, Phenytoin; TST, total sleep time; S4, Stage 4; S2, Stage 2.

M. Corsi-Cabrera et al. / Clinical Neurophysiology 117 (2006) 1826–18351828

second night spent at the laboratory was continuously

captured overnight on a PC computer through a 12-bit

resolution analog-to-digital converter, sampling rate of

512 Hz, with the acquisition program GRASS-GAMMA

version 4.4.

Wakefulness and sleep stages were identified according

to the standardized manual for sleep scoring using 30 s

epochs for scoring (Rechtschaffen and Kales, 1968). The

total sleep time and percentage of stage 2 were within

normal ranges in the two patients (Williams et al., 1974),

while stage 4 was reduced in the patient RC (Table 1).

EEG of stage 2 (S2) and stage 4 (S4) of slow wave sleep

(SWS) and from wakefulness (W) with eyes closed before

going to sleep were carefully inspected for artifacts and

specially for epileptic or abnormal traces. Before going to

sleep, subjects were instructed to close their eyes to

collect waking epochs with eyes closed. Slow (1–4 Hz)

and sleep spindle frequencies are the main landmarks of

SWS and have shown increased coherent activity during

sleep (Corsi-Cabrera et al., 1987; Guevara et al., 1995;

Achermann and Borbely, 1998; Perez-Garci et al., 2001;

Duckrow and Zaveri, 2005). Slow frequencies and sleep

spindles are more prominent during stage 4 and stage 2,

respectively, whereas 1–4 Hz oscillations are not so stable

during stage 3, therefore, stage 2 and stage 4, and not

stage 3, were chosen for analysis. EEG activity free from

artifacts, abnormal activity and eye movements was

segmented into successive non-overlapping 2 s epochs.

A total of 120 2 s epochs were analyzed for each patient

and condition. Correlation spectra at 0 delay for each

patient and physiological state were obtained for 1 Hz

bins from 1 to 25 Hz, using amplitude values by means

of Pearson product-moment coefficients (Guevara and

Corsi-Cabrera, 1996) with the program POTENCOR

(Guevara et al., 2002). This program calculates the Fast

Fourier Transform and then the correlation spectrum is

calculated from the autospectra and from the cross-spectrum

of the signals proportioning values betweenK1 andC1.

Interhemispheric correlation between F3–A1 and F4–A2

and between P3–A1 and P4–A2 derivations were chosen to

study the effect of CC transection because one patient had

anterior 2/3 callosal section and the other complete section

of the CC, thus providing information for cortical regions

(posterior) interconnected by intact CC and cortical regions

(anterior) with sectioned CC, and because the localization of

these electrodes over the replaced bone did not rest upon the

skull gap.

Interhemispheric correlation values were transformed to

Fisher’s z scores with the formula rz Z0.5ln((1Cr)/(1Kr)),

which is usually used to approach correlation values to a

normal distribution before any statistical procedure (Guil-

ford and Fruchter, 1978). The differences between wakeful-

ness and each sleep stage was obtained by subtracting

wakefulness Fisher’s z-transformed correlation values from

corresponding sleep stages (S4 minus wakefulness or S2

minus wakefulness). The data are expressed as the mean

differences between wakefulness and sleep stages. This

method reduces the effect of inter-subject and inter-

electrode pair variability on absolute correlation values

(Hummel and Gerloff, 2005). The significance of the mean

differences between wakefulness and sleep and between

pre- and post-surgical conditions of each patient was

obtained using the formula for correlated groups (Daniel,

1991).

The 95% confidence intervals were established and

mean differences between wakefulness and sleep and

between pre- and post-surgical conditions of each patient

were considered statistically significant when the

confidence intervals of the difference did not reach the

zero mean difference. To determine significant

differences between the control group and pre-surgical

condition of each patient the 95% confidence intervals

were calculated for each condition and the differences

were considered significant if the confidence intervals did

not overlap.

M. Corsi-Cabrera et al. / Clinical Neurophysiology 117 (2006) 1826–1835 1829

3. Results

3.1. Coherent activity before surgery

Mean differences (sleep stage minus wakefulness) in

interhemispheric correlation between wakefulness and sleep

are illustrated in Fig. 1 for stage 4 and in Fig. 2 for stage 2

for the control group and pre-surgical conditions of the two

patients. The significance of the difference between

wakefulness and sleep at 95% confidence intervals are

1.00

0.50

z

0.00

–0.501 3 5 7 9 11

HERTZ

1.00

0.50

z

0.00

–0.50

1.

0.

0.

–0.1 3 5 7 9 11 13 15 17 19 21 23 25

RC

RC

CON

POST

CON

ANTE

STAGE 4 mi

1.00

0.50

z

0.00

–0.501 3 5 7 9 1

HERTZ

1.00

0.50

z

0.00

–0.50

1

0

0

–01 3 5 7 9 11 13 15 17 19 21 23 25

Fig. 1. Each curve connects average Fisher’s z-transformed correlation values exp

the control group and for the patients before surgery for anterior (F3–F4) and pos

values and decrements as negative values. Shaded areas indicate frequency bins

bottom of the figures indicate frequency bins in which the control group and ea

statistical details.

indicated for each frequency bin by shaded areas and

significant differences between the control group and the

patients are indicated by asterisks at the bottom of the

spectra.

3.2. Anterior regions

The control group showed significantly higher coherent

activity during stage 4 at slow frequencies from 1 to 7 Hz,

except for 3 Hz, and in spindle frequencies from 12–16 Hz.

13HERTZ

15 17 19 21 23 25

00

50

z

00

50

HERTZ AS

AS

TROL GROUP

ERIOR REGION

TROL GROUP

RIOR REGION

nus WAKEFULNESS

1 13 15 17 19 21 23 25

z.00

.50

.00

.50

HERTZ1 3 5 7 9 11 13 15 17 19 21 23 25

HERTZ1 3 5 7 9 11 13 15 17 19 21 23 25

ressed as the difference between wakefulness and stage 4 (S4 minus W) for

terior regions (P3–P4). Increments during stage 4 are expressed as positive

in which wakefulness and stage 4 differed significantly and asterisks at the

ch patient differed significantly at a 95% confidence interval. See text for

1.00

0.50

z

0.00

–0.501 3 5 7 9 11 13

HERTZ

HERTZ

15 17 19 21 23 25

1.00

0.50

z

0.00

–0.50

1.00

0.50

z

0.00

–0.501 3 5 7 9 11 13 15 17 19 21 23 25

HERTZRC AS

RC AS

CONTROL GROUP

POSTERIOR REGION

CONTROL GROUP

ANTERIOR REGION

STAGE 2 minus WAKEFULNESS

1.00

0.50

z

0.00

–0.501 3 5 7 9 11 13 15 17 19 21 23 25

HERTZ

1.00

0.50

z z

0.00

–0.50

1.00

0.50

0.00

–0.501 3 5 7 9 11 13 15 17 19 21 23 25

HERTZ1 3 5 7 9 11 13 15 17 19 21 23 25

HERTZ1 3 5 7 9 11 13 15 17 19 21 23 25

Fig. 2. Each curve connects average Fisher’s z-transformed correlation values expressed as the difference between wakefulness and stage 2 (S2 minus W) for

the control group and for the patients before surgery for anterior (F3–F4) and posterior regions (P3–P4). Increments during stage 2 are expressed as positive

values and decrements as negative values. Shaded areas indicate frequency bins in which wakefulness and stage 2 differed significantly and asterisks at the

bottom of the figures indicate frequency bins in which the control group and each patient differed significantly at a 95% confidence interval. See text for

statistical details.

M. Corsi-Cabrera et al. / Clinical Neurophysiology 117 (2006) 1826–18351830

A generalized increase in coherent activity during S4

compared to W was appreciated before surgery in almost all

frequencies of the spectra in the two patients (except for

some isolated frequencies: 1, 2, 4, 5, 6, 12, and 15 HZ in

patient RC and 1–7, 13, and 16 Hz in patient AS) including

those within the range of alpha activity, where there were no

significant differences in the control group, and in beta

frequencies where the control group showed a decrease.

Both patients showed higher increase than the control group

at all frequencies except patient AS at 13 and 14 Hz where

the increase is similar to the control group.

During stage 2 the control group showed increased

coherent activity for slow frequencies from 1 to 6 Hz except

for 3 Hz and from 13 to 16 Hz. Coherent activity in patient

RC was higher during stage 2 than during wakefulness at

10–12, 16 and some frequencies between 19 and 25 Hz.

Patient AS showed higher coherent activity during stage 2

compared to wakefulness at 11 and 12 Hz. The increase

M. Corsi-Cabrera et al. / Clinical Neurophysiology 117 (2006) 1826–1835 1831

from wakefulness to stage 2 was larger in patient RC, while

in the patient AS it was lower than in the control group.

3.3. Posterior regions

Coherent activity between posterior regions was signi-

ficantly higher during stage 4 compared to wakefulness

from 1 to 16 Hz in the control group and in the two patients

(except for 5 Hz in the control group, 1, 5, and 12–15 Hz in

RC and for 4, 7, and 8 Hz in AS). The increase was larger in

both patients compared to the control group in some slow,

alpha and beta frequencies, while in spindle frequencies it

was smaller in RC and similar to the control group in AS.

Given the peak in coherent activity in alpha frequencies

in both patients, polysomnographic traces of S2 and S4 from

EEG epochs close to those used for quantitative analysis

were visually inspected to look for alpha rhythm. No alpha

was detected upon visual inspection, whereas a reduction in

intraspindle frequency was appreciated in sleep spindles

with typical morphology and duration.

1.00

0.50

z

0.00

–0.501 3 5 7 9 11 13

HERTZ15 17 19 21 23 25

1

0

0

–0

1.00

0.50

z

0.00

–0.501 3 5 7 9 11 13

HERTZ15 17 19 21 23 25

1

0

0

–0

ANTERIOR R

POSTERIOR R

RC

RC

STAGE 4 minus WA

Fig. 3. Each curve connects average Fisher’s z-transformed correlation values exp

patient RC after full callosotomy and for patient AS after anterior 2/3 callosotomy

4 are expressed as positive values and decrements as negative values. Shaded

significantly and asterisks at the bottom of the figures indicate frequency bins in wh

a 95% confidence interval. Interhemispheric black lines in the insets illustrate ca

3.4. Coherent activity after surgery

3.4.1. Anterior regions

Changes in coherent activity from wakefulness to sleep

after surgery are shown in Fig. 3 for S4 and in Fig. 4 for S2.

The significance of the difference between wakefulness and

sleep at the 95% confidence intervals are indicated for each

frequency bin by shaded areas and between pre- and post-

surgery conditions by asterisks at the bottom of the spectra.

Coherent activity after callosotomy was significantly

higher during stage 4 than during wakefulness at 10 and

16 Hz and from 20 to 25 Hz in RC, and at 7, 9, and 10 Hz

and at all frequencies from 12 to 25 Hz in AS. However, the

increase was smaller than before surgery in almost all

frequencies.

Coherent activity was also higher during stage 2 than

during wakefulness from 9 to 25 Hz in patient RC but

smaller than before surgery in all frequencies except at

13 Hz, where it was higher than before surgery. The

increase in coherent activity from wakefulness to sleep

was not significant in patient AS.

.00

.50

z

.00

.501 3 5 7 9 11 13

HERTZ15 17 19 21 23 25

.00

.50

z

.00

.501 3 5 7 9 11 13

HERTZ15 17 19 21 23 25

EGION

EGION

AS

AS

KEFULNESS

ressed as the difference between wakefulness and stage 4 (S4 minus W) for

for anterior (F3–F4) and posterior regions (P3–P4). Increments during stage

areas indicate frequency bins in which wakefulness and stage 4 differed

ich pre- and post-surgery conditions in each patient differed significantly at

llosotomy region. See text for statistical details.

STAGE 2 minus WAKEFULNESS

ANTERIOR REGIONRC AS

RC AS

1.00

0.50

0.00

–0.50

1.00

0.50

0.00

–0.50

1.00

0.50

0.00

–0.50

1.00

0.50

0.00

–0.501 3 5 7 9 11 13

HERTZ

15 17 19 21 23 25

1 3 5 7 9 11 13

HERTZ

15 17 19 21 23 25

1 3 5 7 9 11 13

HERTZ

15 17 19 21 23 25

1 3 5 7 9 11 13

HERTZ

15 17 19 21 23 25

Z Z

Z Z

Fig. 4. Each curve connects average Fisher’s z-transformed correlation values expressed as the difference between wakefulness and stage 2 (S2 minus W) for

patient RC after full callosotomy and for patient AS after anterior 2/3 callosotomy for anterior (F3–F4) and posterior regions (P3–P4). Increments during stage

2 are expressed as positive values and decrements as negative values. Shaded areas indicate frequency bins in which wakefulness and stage 2 differed

significantly and asterisks at the bottom of the figures indicate frequency bins in which pre- and post-surgery conditions in each patient differed significantly at

a 95% confidence interval. Interhemispheric black lines in the insets illustrate callosotomy region. See text for statistical details.

M. Corsi-Cabrera et al. / Clinical Neurophysiology 117 (2006) 1826–18351832

The large alpha coherent peak was reduced after surgery

in both patients, while the increase in beta frequencies

remained in RC and disappeared in AS.

3.4.2. Posterior regions

The effect of surgery on the interhemispheric coherent

activity increase during S4 in patient RC with complete

callosotomy and in patient AS with intact CC is shown in

the low part of Fig. 3. Coherent activity was higher during

stage 4 than during wakefulness in patient RC from 1 to 7

and from 13 to 15 Hz, and in almost all frequencies except

for 1, 3, 8, and 10 Hz in patient AS. The increase in coherent

activity during stage 4 was significantly larger in slow

frequencies and lower in fast frequencies than before

surgery in RC except for 4 and 13 Hz where it was similar

to pre-surgical levels despite callosotomy. The increase

from wakefulness to stage 4 was also reduced in patient AS

regardless of the integrity of CC in almost all frequencies

except for 1, 3, 8, 10, and 24 Hz where it was similar to pre-

surgical levels.

There was a significant increase from wakefulness to

stage 2 in patient RC in almost all frequencies except for

isolated frequencies and in patient AS at 3, 11, 13, and

15 Hz. The increase of coherent activity from wakefulness

to stage 2 was also reduced after surgery in both patients in

beta frequencies, while it was larger in many other

frequencies regardless of the section of the CC in one of

them.

4. Discussion

The two patients showed higher coherent activity in

specific sleep-related frequencies during sleep compared to

wakefulness after surgery. If the increase in interhemi-

spheric coherent activity during sleep were only dependent

on the connection between left and right hemispheres via the

CC, suppression of this increase should be clearly observed

both in S2 and S4 with respect to W, between anterior

regions in the two patients and between posterior regions

M. Corsi-Cabrera et al. / Clinical Neurophysiology 117 (2006) 1826–1835 1833

exclusively in the patient who underwent complete

callosotomy. Results observed in both patients also indicate

that the CC plays an important role in the increase of

interhemispheric coherent activity during sleep, since it was

attenuated in some frequencies after callosotomy. The

decrease in coherent activity after callosotomy is consistent

with lower coherent activity during sleep compared to

control subjects after callosotomy (Montplaisir et al., 1990)

and in Alzheimer patients with reduced CC volume

(Pogarell et al., 2005). However, these results also indicate

that the CC is not the only factor underlying the increase in

coherent activity during sleep because coherent activity was

higher during sleep than during wakefulness after callo-

sotomy in many frequencies of the spectrum, particularly

because it was also modified between posterior regions in

the patient with only anterior callosotomy and remained the

same in some frequencies in the patient with complete

callosotomy.

The persistence of interhemispheric coherent activity

after CC section is consistent with a previous study of one

case with anterior 2/3 callosotomy where we found that

interhemispheric coherent activity during wakefulness

1 year after callosotomy was similar (overlapping the

first standard deviation) to a normal group of women

(Corsi-Cabrera et al., 1995). Thus, other factors besides the

CC, such as reduction of interfering activity during normal

sleep and common inputs influencing both, frontal and

parietal cortical regions that may also influence coherent

activity have to be responsible for the increase in interhemi-

spheric coherent activity from wakefulness to sleep.

W and SWS are characterized by distinctive firing

patterns of cortical neurons resulting in correlated field

potentials and EEG oscillatory activity, delta and sleep

spindles in SWS, and a predominance of fast activity in the

range of beta and gamma frequencies in W (Steriade, 1998).

The generation of the distinct firing patterns of cortical

neurons is possible due to their intrinsic membrane

properties, and the shift from one pattern to another

according with the physiological state is produced by direct

and indirect influences from systems involved in the

generation of arousal states (Steriade, 1998). The simul-

taneous occurrence of firing patterns of cortical neurons and

the correlated EEG activity all over the cortex may depend

on reciprocal cortico–cortical connections (Edelman and

Tononi, 2000), but also on common subcortical afferents

affecting the cortex. There is strong evidence for the role of

the thalamo–cortico–thalamic network in the spread of the

same pattern of oscillatory activity over cortical regions.

The simultaneous occurrence of spindles in natural SWS is

not affected by the interruption of intracortical pathways in

the cat (Contreras et al., 1996); thus, the increase in coherent

activity during sleep compared to wakefulness may also

depend on the influences of the thalamo–cortico–thalamic

network during sleep.

Present results have to be taken with caution, first of all

because they come from only two cases and therefore they

cannot be extended to the general population without further

investigation, and second, because they come from patients

with a long history of epilepsy and polypharmacology. The

EEG epochs that entered the analysis were free of any sign

of epileptic or abnormal traces, nevertheless coherent

activity during sleep before surgery showed some striking

differences compared to the control group indicating

possible alterations in background electrical activity due

to plastic changes caused by both epilepsy and/or its

correlated therapy. The increase in coherent activity from

wakefulness to SWS in slow and in spindle frequencies was

larger than in the control group and it was also higher in

alpha and beta frequencies where usually it is not observed.

The large increase in coherent activity in alpha frequencies

was evident in both stage 2 and stage 4 of SWS and was

larger between frontal than between parietal derivations.

Visual inspection of polysomnographic traces indicated no

arousal related alpha activity and abundant sleep spindles

with low intraspindle frequencies in S2 and S4. These

findings are consistent with results reported in the literature

for two types of sleep spindle activity, a slower one of about

12 Hz predominant over frontal regions and a faster one of

about 14 Hz predominant over parietal regions (Nakamura

et al., 2003; Jobert et al., 1992) and with coherence peaks at

8 and 10 Hz during SWS found in some subjects (Duckrow

and Zaveri, 2005). They also agree with evidence of higher

incidence of 10 Hz spindles in patients with complex partial,

and partial secondarily generalized seizures (Drake et al.,

1991) and even lower intraspindle frequencies (Sengoku

and Wolf, 1981). Thus, the large peak of increased coherent

activity in alpha frequencies is more probably reflecting

increased coherent activity in low frequency spindles. The

focus of the study was to investigate the influence of the

integrity of the CC on the increase of interhemispheric

coherent activity from wakefulness to sleep, which is

essentially provided by pre- and post-surgical differences

between wakefulness and sleep in the two patients, and not

to derive general physiological conclusions nor to charac-

terize sleep before and after callosotomy. Sleep EEG and

architecture is being reported in detail elsewhere (in

preparation).

Coherence between two brain regions, especially from

scalp recordings, contains contribution from both neural

activity and common input to the electrodes by volume

conduction of current as a result, coherence may be strongly

inflated by volume conduction at short distances and at low

frequencies (Leopold and Logothetis, 2003), while volume

conductivity can be negligible between long distances

(Nunez et al., 2001); the distance between left and right

frontal and left and right parietal electrodes used for the

comparisons in the present study is larger than 12 cm

minimizing volume conduction effects; therefore, the

reported changes after surgery for anterior (frontal) and

posterior (parietal) regions may be attributed to callosotomy

and not only to volume conduction. Volume conduction

effects on correlation values cannot be completely ruled out;

M. Corsi-Cabrera et al. / Clinical Neurophysiology 117 (2006) 1826–18351834

however, significant differences between pre- and post-

surgical conditions were not based on absolute correlation

values which may be inflated by volume conduction, but on

the difference between wakefulness and sleep by subtracting

wakefulness values from corresponding sleep stages. This

method reduces the effect of inter-subject and inter-

electrode pair variability and minimizes the effect intro-

duced by the reference electrodes on absolute correlation

values (Hummel and Gerloff, 2005). Sleep stage, frequency

bin, and electrode pair was the same in each comparison.

There is no evidence until now that the physiological state,

W, S2, and S4, could modify volume conduction for the

same frequency and for the same electrode pair. Thus,

similar effects of volume conduction and common inputs to

the electrodes should be expected for the same frequency

and electrodes combination. The only difference in each

comparison was the interruption and/or integrity of callosal

fibers. Post-surgical recordings were done 4 months after

surgery, avoiding the influence of acute surgical effects on

EEG activity. The localization of parietal and frontal

electrodes did not overlap skull gaps and in any case both

wakefulness and sleep EEG would be affected in a similar

way by skull opening effects.

Regarding the effects of antiepileptic drugs on interhemi-

spheric coherent activity, there is no information available

in the literature that could help to exclude pharmachological

effects on EEG coherent activity. Present results open the

question of a possible effect of pharmacological treatment

on coherent activity in epileptics, which is presently being

investigated in the laboratory. The increase in coherent

activity in the beta band was evident in S2 and S4, and

remained elevated after callosotomy. We have shown

increased interhemispheric coherent activity in beta

frequencies with diazepam (Romano-Torres et al., 2002),

which acts on GABAA receptors increasing their inhibitory

action. A possibility exists, that needs to be tested, that

anticonvulsant drugs may have a similar effect on EEG

coherent activity explaining the larger increase in coherent

activity in these frequencies and/or in slow and spindle

frequencies compared to control subjects.

Although present results come from two patients only,

they clearly demonstrate that interhemispheric coherent

activity, despite an attenuation after surgery, is higher

during SWS than during wakefulness after sectioning the

CC, showing that the higher coherent activity during sleep

does not depend exclusively on callosal connections but also

on state-dependent influences of sleep promoting

mechanisms probably spread through the thalamo-cortical

network.

Acknowledgements

Isabel Perez-Montfort corrected the English version of

the manuscript. This work was partially financed by

DGAPA, project IN214702.

References

Achermann P, Borbely AA. Coherente analisis of the human sleep

electroencephalogram. Neuroscience 1998;85:1195–208.

Contreras D, Destexhe A, Sejnowski TJ, Steriade M. Control of

spatiotemporal coherence of a thalamic oscillation by corticothalamic

feedback. Science 1996;274:771–4.

Corsi-Cabrera M, Meneses S, Molina E. Correlacion interhemisferica y

Acoplamiento temporal de la actividad electrica cortical durante la

vigilia, la etapa II y el sueno paradojico en el hombre. Rev Mex Psicol

1987;4:100–8.

Corsi-Cabrera M, Ramos J, Meneses S. Effect of normal sleep and sleep

deprivation on interhemispheric correlation during subsequent wakeful-

ness in man. Electroencephalogr Clin Neurophysiol 1989;72:305–11.

Corsi-Cabrera M, Trıas G, Guevara MA, Haro R, Hernandez A. EEG

interhemispheric correlation after callosotomy: one case study. Percept

Mot Skills 1995;80:504–6.

Corsi-Cabrera M, Guevara MA, Arce C, Ramos J. Inter and intrahemi-

spheric EEG correlation as a function of sleep cycles. Prog

NeuroPsychopharmacol Biol Psychiatry 1996;20:387–405.

Corsi-Cabrera M, Munoz-Torres Z, del Rıo Portilla Y, Guevara MA. Power

and coherent oscillations distinguish REM sleep, stage 1 and

wakefulness. Int J Psychophysiol 2006;60(1):59–66.

Crick F. The astonishing hypothesis: the scientific search for the soul. New

York: Touchstone; 1994.

Daniel W. Biostatistics: a foundation for analysis in the health sciences.

Somerset, NJ: John Wiley & Sons; 1991.

Drake Jr ME, Pakalnis A, Padamadan H, Weate SM, Cannon PA. Sleep

spindles in epilepsy. Clin Electroencephalogr 1991;22:144–9.

Duckrow RB, Zaveri HP. Coherence of the electroencephalogram during

the first sleep cycle. Clin Neurophysiol 2005;16(5):1088–95.

Edelman GM, Tononi G. A universe of consciousness. New York: Basic

Books; 2000.

Fuchigami T, Mazaki R, Nishimura A, Noguchi Y, Fuchigami S, Fujita Y,

Okubo O, Harada K. A mother and daughter with agenesis of the corpus

callosum. Acta Paediatr Jpn 1996;38:52–6.

Gais S, Born J. Declarative memory consolidation: mechanisms acting

during human sleep. Learn Mem 2004;11(6):679–85.

Gray M. The temporal correlation hypothesis of visual feature integration:

still alive and well. Neuron 1999;24:31–47.

Guevara MA, Corsi-Cabrera M. EEG correlation or EEG coherence? Int

J Psychophysiol 1996;23:145–53.

Guevara MA, Lorenzo I, Arce C, Ramos J, Corsi-Cabrera M. Inter- and

intrahemispheric EEG correlation during sleep and wakefulness. Sleep

1995;18:257–65.

Guevara MA, Ramos J, Hernandez-Gonzalez M, Zarabozo D, Corsi-

Cabrera M. POTENCOR: a program to calculate power and correlation

spectra of EEG signals. Comput Methods Programs Biomed 2002;72:

241–50.

Guilford JP, Fruchter B. Fundamental statistics in psychology and

education. New York: McGraw-Hill; 1978.

Hummel F, Gerloff Ch. Larger interregional synchrony is associated with

greater behavioral success in a complex sensory integration task in

humans. Cereb Cortex 2005;15:670–8.

Jobert M, Poiseau E, Jahnig P, Schulz H, Kubicki S. Topographical analysis

of sleep spindle activity. Neuropsychobiology 1992;26:210–7.

Kuks JB, Vos JE, O’Brien MJ. Coherence patterns of the infant sleep EEG

in absence of the corpus callosum. Electroencephalogr Clin Neurophy-

siol 1987;66:8–14.

Laureys S, Peigneux P, Phillips C, Fuchs S, Degueldre C, Aerts J, Del

Fiore G, Petiau C, Luxen A, Van der Linden M, Cleeremans A,

Smith C, Maquet P. Experience-dependent changes in cerebral

functional connectivity during human rapid eye movement sleep.

Neuroscience 2001;105:521–5.

Leopold DA, Logothetis NK. Spatial patterns of spontaneous local field

activity in the monkey visual cortex. Rev Neurosci 2003;14:195–205.

M. Corsi-Cabrera et al. / Clinical Neurophysiology 117 (2006) 1826–1835 1835

Maquet P, Laureys S, Peigneux P, Fuchs S, Petiau C, Phillips C, Aerts J, Del

Fiore G, Degueldre C, Meulemans T, Luxen A, Franck G, Van del

Linden M, Smith C, Cleeremans A. Experience-dependent changes in

cerebral activation during human REM sleep. Nat Neurosci 2000;3:

831–6.

Marino R, Cukiert A, Gronich G. Open and stereotactic segmental

callosotomy: effects on seizure frequency. In: Reeves AG,

Roberts DW, editors. Epilepsy and the corpus callosum. New York:

Plenum Press; 1995. p. 209–16.

McIntyre DC. Forebrain commissures and limbic kindling. In: Reeves AG,

Roberts DW, editors. Epilepsy and the corpus callosum. New York:

Plenum Press; 1995. p. 79–89.

Molle M, Marshall L, Gais S, Born J. Learning increases human

electroencephalographic coherence during subsequent slow sleep

oscillations. Proc Natl Acad Sci USA 2004;101:13963–8.

Montplaisir J, Nielsen T, Cote J, Boivin D, Rouleau I, Lapierre G.

Interhemispheric EEG coherence before and after partial callosotomy.

Clin Electroencephalogr 1990;21:42–7.

Nakamura M, Uchida S, Maehara T, Kawai K, Hirai N, Nakabayashi T,

Arakaki H, Okubo Y, Nishikawa T, Shimizu H. Sleep spindles in human

prefrontal cortex: an electrocorticographic study. Neurosci Res 2003;

45:419–27.

Nielsen TA, Abel A, Lorrain D, Montplaisir J. Interhemispheric EEG

coherence during sleep and wakefulness in left and right-handed

subjects. Brain Cogn 1990;14:113–25.

Nielsen T, Montplaisir J, Lassonde M. Decreased interhemispheric EEG

coherence during sleep in agenesis of the corpus callosum. Eur Neurol

1993;33:173–6.

Nunez PL, Wingeier BM, Silberstein RB. Spatial-temporal structures of

human alpha rhythms: theory, microcurrent sources, multiscale

measurements and global binding of local networks. Hum Brain

Mapp 2001;13:125–64.

Perez-Garci E, del Rıo-Portilla Y, Guevara MA, Arce C, Corsi-Cabrera M.

Paradoxical sleep is characterized by uncoupled gamma activity

between frontal and perceptual cortical regions. Sleep 2001;24:118–26.

Pogarell O, Teipel SJ, Juckel G, Gootjes L, Moller T, Burger K,

Leinsinger G, Moller HJ, Hegerl U, Hampel H. EEG coherence reflects

regional corpus callosum area in Alzheimer’s disease. J Neurol

Neurosurg Psychiatry 2005;76(1):109–11.

Rechtschaffen A, Kales A. A manual of standardized terminology,

techniques and scoring system for sleep stages of human subjects.

Brain Information Service, Brain Research Institute, University of

California, Los Angeles; 1968.

Romano-Torres M, Borja-Lascurain E, Chao-Rebolledo C, del-Rıo-

Portilla Y, Corsi-Cabrera M. Effect of diazepam on EEG power and

coherent activity: sex differences. Psychoneuroendocrinology 2002;27:

821–33.

Schwartz ML. Organization and development of interhemispheric

connections of the prefrontal cortex in rhesus monkey. In:

Reeves AG, Roberts DW, editors. Epilepsy and the corpus callosum

II. New York: Plenum Press; 1995. p. 29–48.

Sengoku A, Wolf P. Spindles and spike waves in the sleep EEG of epileptic

patients under administration of phenobarbital (PB) and phenytoin

(DPH) (author’s transl). EEG EMG Z Elektroenzephalogr Elektro-

myogr Verwandte Geb 1981;12(3):132–4.

Singer W. Neuronal synchrony: a versatile code for the definition of

relations? Neuron 1999;24:49–65.

Steriade M. Corticothalamic networks, oscillations and plasticity. In: Jasper

HH, Descarries L, Castellucci VF, Rossignol L, editors. Consciousness:

at the frontier of neuroscience. Adv Neurol 1998;77:105–34.

Tononi G, Sporns O, Edelman GM. Reentry and the problem of integrating

multiple cortical areas: simulation of dynamic integration in the visual

system. Cereb Cortex 1992;2:310–35.

Vyazovskiy V, Achermann P, Borbely AA, Tobler I. Interhemispheric

coherence of the sleep electroencephalogram in mice with congenital

callosal dysgenesis. Neuroscience 2004;124(2):481–8.

Wada JA. Midline subcortical structures for transhemispheric ictal and

interictal transmission. In: Reeves AG, Roberts DW, editors.

Epilepsy and the corpus callosum II. New York: Plenum Press;

1995. p. 61–78.

Williams RL, Karacan I, Hursch CJ. Electroencephalography (EEG) of

human sleep: clinical applications. New York: Wiley; 1974.

Williamson PD. Corpus callosum section: preoperative evaluation. In:

Reeves AG, Roberts DW, editors. Epilepsy and the corpus callosum.

New York: Plenum Press; 1995. p. 123–36.

Wilson MA, McNaughton BL. Reactivation of hippocampal ensamble

memories during sleep. Science 1994;265:676–9.